A fluorescent rosamine compound selectively stains pluripotent stem cells

Article abstract of DOI:10.1002/anie.201002463

The cell-by date: The fluorescent compound CDy1 selectively stains embryonic stem cells (see scheme). CDy1 was used to identify fibroblasts undergoing reprogramming to induced pluripotent stem cells prior to the expression of green fluorescent protein under the control of the Oct4 promoter.

Full text of DOI:10.1002/anie.201002463

Angewandte
Chemie
DOI: 10.1002/anie.201002463
Imaging Agents
A Fluorescent Rosamine Compound Selectively Stains Pluripotent
Stem Cells**
Chang-Nim Im, Nam-Young Kang, Hyung-Ho Ha, Xuezhi Bi, Jae Jung Lee, Sung-Jin Park,
Sang Yeon Lee, Marc Vendrell, Yun Kyung Kim, Jun-Seok Lee, Jun Li, Young-Hoon Ahn,
Bo Feng, Huck-Hui Ng, Seong-Wook Yun,* and Young-Tae Chang*
Stem cells, which are capable of self-renewing and differ-
entiating into various types of cells, have captured great
interest as a valuable resource for regenerative medicine and
developmental biology research. Technical progress during
the last decade has enabled the isolation of stem cells from a
wide range of tissues, their differentiation into specific types
of cells, and the generation of induced pluripotent stem cells
(iPSC) from somatic cells. The recent success of patient-
specific iPSC generation[1] and their differentiation into
functional cells[2] exemplifies how stem cells can be used for
drug discovery and treatment of specific patients having
complex diseases.[2]
However, despite the general enthusiasm for the multiple
applications of stem cells, their practical use both in research
and disease therapy has been hampered by the heterogeneity
of stem cells and their unpredictable proliferation and
differentiation.[3] The current methods of isolation and
characterization of stem cells mostly depend on their
morphology in the culture, such as colony or sphere forma-
tion, and immunodetection of marker proteins. These meth-
ods, however, require extended times and antibody reactions
which may make the cells unsuitable for further usage.
Therefore, the development of tools and technologies that
may facilitate the isolation, identification, and characteriza-
tion of stem cells is one of the most demanding requisites in
the field of stem-cell research and applications.
[*] Dr. C.-N. Im,[+] Dr. H.-H. Ha, Dr. S. Y. Lee, Dr. Y. K. Kim, Dr. J.-S. Lee,
Prof. Dr. Y.-T. Chang
Department of Chemistry, National University of Singapore
3 Science Drive 3, Singapore 117543 (Republic of Singapore)
Fax: (+65)6779-1691
E-mail: chmcyt@nus.edu.sg
Homepage: http://ytchang.science.nus.edu.sg
Dr. N.-Y. Kang,[+] Dr. X. Bi, Dr. J. J. Lee, Dr. S.-J. Park, Dr. M. Vendrell,
Dr. J. Li, Dr. S.-W. Yun, Prof. Dr. Y.-T. Chang
Laboratory of Bioimaging Probe Development
Fluorescent small molecules have been widely used for
the visualization of polymeric biomolecules or cellular
organelles.[4] We have employed combinatorial chemistry to
develop several diversity-oriented fluorescence libraries
(DOFL) and successfully applied them to the discovery of
imaging probes for a number of biological targets.[5] Among
our libraries is a rosamine library synthesized using solid-
phase chemistry[6] to provide more flexibility within the
rhodamine scaffold, which has excellent photophysical prop-
erties. By screening this library against a muscle-formation
cell culture, we previously discovered a compound that
controls muscle differentiation.[7] To additionally evaluate
the application of rosamine derivatives as stem-cell-selective
probes, we have screened the library against embryonic stem
cells (ESC) in this study and discovered a novel fluorescent
compound, called the compound of designation yellow 1
(CDy1, lex/lem = 535/570 nm), that selectively stains ESC and
iPSC as well.
For high-throughput screening, we incubated mouse ESC
(mESC) and mouse embryonic fibroblast (MEF) feeder cells
with 280 rosamine compounds at a concentration of 500 nm in
384-well microplates. After 0.5, 24, and 48 hours, tetrame-
thylrhodamine isothiocyanate (TRITC) fluorescence and
bright-field images were taken using an ImageXpressMICRO
imaging system. From this primary screening, 20 compounds
that stained mESC consistently with stronger intensity than
MEF were manually selected. As a secondary screening, we
incubated mESC and MEF separately with each of the hit
compounds and analyzed them using flow cytometry and
Singapore Bioimaging Consortium, Agency for Science
Technology and Research (A*STAR), 11 Biopolis Way
# 02-02 Helios, Singapore 138667 (Republic of Singapore)
E-mail: yun_seong_wook@sbic.a-star.edu.sg
Dr. H.-H. Ha, Prof. Dr. Y.-T. Chang
NUS MedChem Program of Life Sciences Institute, National
University of Singapore, Singapore 117543 (Republic of Singapore)
Dr. Y. K. Kim, Dr. J.-S. Lee, Dr. Y.-H. Ahn
Department of Chemistry, New York University
New York, NY 10003 (USA)
Dr. B. Feng, Dr. H.-H. Ng
Gene Regulation Laboratory, Genome Institute of Singapore
Agency for Science, Technology and Research (A*STAR)
Singapore 138672 (Republic of Singapore)
Dr. H.-H. Ng
Department of Biological Sciences, National University of Singa-
pore, Singapore 117543 (Republic of Singapore)
[+] These authors contributed equally to this work.
[**] We thank Sai Kiang Lim (Institute of Medical Biology, Singapore) for
providing mESC and MEF and kind consultation. We also thank Siti
Hajar, Chew Yan Tuang, Yee Ling Tan, Pei Fen Tay, and Si Qiang Yang
for excellent technical support in cell culture and screening, and
Clement Khaw (SBIC-Nikon Imaging Centre) for microscopy. This
work was supported by a Young Investigator Award (R-143-000-353-
101) granted to Y.-T.C. from the National University of Singapore
and intramural funding from the A*STAR Biomedical Research
Council. J.-S.L. was supported by a Korea Research Foundation
Grant funded by the Korean government (MOEHRD, Basic
Research Promotion Fund: KRF-2005-C00088).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002463.
Angew. Chem. Int. Ed. 2010, 49, 7497 –7500
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7497

Communications
found CDy1 to be the most selective for mESCs among the 20
(Figure 1d). RT-PCR analysis clearly showed that Nanog
gene expression in SSClow CDy1bright cells is much higher than
that in SSChigh CDy1dim cells (Figure 1e), and the numbers of
colonies counted after a three-day culture were 604 and 6,
respectively. We additionally induced differentiation of these
CDy1-stained mESC by culturing in media without the
leukemia inhibitory factor (LIF). After two weeks of differ-
entiation, the expression of ectoderm, mesoderm, and endo-
derm markers were verified by immunocytochemistry (see
Figure S4 in the Supporting Information). These data dem-
onstrate that CDy1 can be used for mESC enrichment from a
mixed-cell population without affecting the properties of the
stem cells.
hit compounds (Figure 1a–c; see Schemes S1 and S2 in the
Supporting Information). For a more systematic structure–
selectivity relationship study, we synthesized a number of
Having found that CDy1 selectively stains ESC, we
applied the dye to iPSC which was generated from MEF of
transgenic mice that express green fluorescent protein (GFP)
under the control of the Oct4 (also known as Pou5f1)
promoter. The reprogramming was performed in a 6-well
culture dish by retroviral introduction of four transcriptions
factors, Oct4, Sox2, Klf4, and c-Myc,[8] and iPSC generation
was verified by GFP expression, an alkaline phosphatase
assay, and immunostaining of SSEA-1 at 17 days post
infection (dpi). We found CDy1 also selectively stains the
iPSC colony (see Figure S5 in the Supporting Information).
When the 155 colonies grown in a 6-well plate cells were
treated with CDy1 at 17 dpi, 101 colonies (65%) were both
CDy1 and GFP positive, 26 (17%) were CDy1-only positive,
4 (3%) were GFP-only positive and 24 (15%) were negative
for both CDy1 and GFP, despite the fact that the morphology
of the colonies was indistinguishable (see Figure S6 in the
Supporting Information). In a cell culture treated with CDy1
at an earlier time point of iPSC generation (10 dpi), an
increasing number of CDy1-stained colonies started to show a
GFP signal (Figure 2a). To perform a more systematic
analysis, we stained the cells with CDy1 at 2 dpi, when iPSC
was not distinguishable by any means, and tracked the CDy1
and GFP signals by daily acquisition of cell images using an
ImageXpressMICRO system. At 10 dpi, when small colonies
started to appear, we selected 342 CDy1-positive colonies that
were not expressing GFP and monitored their GFP expres-
sion up to 25 dpi. More and more colonies started to express
GFP during this period, and 338 (99%) out of the 342 tracked
colonies were GFP positive at 25 dpi (Figure 2b). During this
period any detectable differences in the number of GFP-
positive colonies or cell morphology were not observed
compared to untreated iPSC. In contrasnt, we induced mESC
differentiation by removing the LIF from the culture media
and observed some cells that were morphologically distin-
guishable from mESC 3 days later. Most of the differentiated
cells were not stained by CDy1, whereas some other cells
having mESC morphology were stained by the dye, and
showed similar patterns for immunocytochemical staining
using the Oct4 antibody (Figure 3). This result was addition-
ally confirmed in lineage-specific cells differentiated from
mESC by embryoid body formation (see Figure S4 in the
Supporting Information).
Figure 1. Selective staining of mESC by CDy1. a) Chemical structure of
CDy1. b) Upper panel: CDy1-stained mESC were immunostained with
anti-Oct4 antibody; lower panel: DMSO was used as a negative
control. c) Flow cytometry dot-plot image of CDy1-stained mESC and
MEF. Images of pure cell populations were overlaid. d) Flow cytometry
dot-plot image of mESC and MEF mixture. Upper panel: mESC and
MEF mixed cells incubated with DMSO; lower panel: mESC and MEF
mixed cells incubated with CDy1. e) Nanog gene expression analysis
using RT-PCR. SSClow CDy1bright and SSChigh CDy1dim cells were sorted
from a mESC and MEF mixture after CDy1 staining. BF=bright field;
scale bar: 100 mm.
CDy1 analogues by modifying the morpholine group (see
Scheme S1 in the Supporting Information) because most of
the 20 primary hits had a di-n-butyl group in common. The
mESC selectivity of all the analogues were, however, much
lower than CDy1 which showed 12.2-fold higher intensity in
mESC than in MEF. This result suggests that the morpholine
group is also important for ESC selectivity of CDy1 (see
Table S1 and Figure S3 in the supporting Information).
To evaluate the capability of CDy1 to isolate ESC from a
mixed-cell population, we stained the MEF and mESC
mixture with CDy1, gated the mixed cells into side scatter
(SSC)low CDy1bright and SSChigh CDy1dim regions, and collected
40000 cells using a fluorescence-activated cell sorter (FACS)
from each gate for analysis of a stem-cell marker Nanog gene
expression and a colony-forming assay. The dot-plot image of
the CDy1-stained mixed cells was similar to the overlay image
of pure populations, whereas the cells incubated with
dimethyl sulfoxide, used as a control, were not distinguishable
In our previous study with the rosamine library,[7] CDy1
was among the compounds targeting mitochondria. To
examine if CDy1 localizes in mitochondria in mESC, we co-
7498 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7497 –7500

Angewandte
Chemie
Figure 4. Co-localization of CDy1 and MitoTracker. mESC were incu-
bated with 500 nm CDy1, 200 nm MitoTracker Deep Red 633, and
4 mgmLÀ1 Hoechst for 30 min at 378C, and then washed with PBS
(pH 7.4) two times. The cells were mounted on a glass slide and the
images were recorded using a Nikon ECLIPSE Ti fluorescence micro-
scope equipped with a ꢀ100 objective lens. PBS=phosphate-buffered
saline. Pearson’s correlation coefficient=0.88; scale bar: 5 mm.
known to express a high level of ALDH1A1 and is stained by
Aldefluor[11] (Figure 5). The stemness of hESC BG01V used
in this study was verified by immunocytochemical staining of
TRA-1-60 (see Figure S7 in the Supporting Information).
Figure 2. iPSC staining by CDy1 at an early stage of reprogramming.
a) iPSC selective staining by CDy1. At 10 days post retroviral infection
(dpi) with Oct4, Sox2, Klf4, and c-Myc, the iPSCs generated from Oct4-
GFP transgenic mouse MEF were stained with CDy1. The pictures of
the same colony were taken at 11 and 12 dpi. b) Time-course analysis
of CDy1-stained colonies. Among the 342 CDy1-positive colonies that
were not expressing GFP counted at 10 dpi, 338 colonies expressed
GFP at 25 dpi. Scale bar: 200 mm
Figure 3. Differentiated mESC staining with CDy1. Morphologically
distinguishable differentiated cells were observed after a three-day
culture of mESC in the absence of LIF. Most of those cells were CDy1
negative, whereas some other cells that retained mESC morphology
were stained by the dye. Immunocytochemistry with the Oct4 antibody
performed on the following day showed a similar pattern of staining
CDy1. Scale bar: 100 mm.
Figure 5. Comparison of CDy1 and Aldefluor staining in mESC and
hESC (BG01 V). Aldefluor does not stain ESC, whereas CDy1 stains
ESC but not the human lung cancer cell line H522, which expresses
ALDH1a1. The cells cultured on a six-well plate were incubated with
1uM Aldefluor for 1 h at 378C and then washed with PBS three times
before the Aldefluor assay buffer was added. Bright-field and fluores-
cence (FITC filter for Aldefluor and TRITC filter for CDy1) images were
recorded using a Nikon ECLIPSE Ti fluorescence microscope. FITC=
fluorescein isothiocyanate. Scale bars: 100 mm.
stained the cells with CDy1 and a mitochondria-staining
commercial dye (MitoTracker Deep Red 633) and observed a
CDy1 staining pattern that was very similar to that of the
MitoTracker staining (Figure 4). In addition to the mitochon-
drial membrane potential which sequesters many cationic
rhodamine and rosamine compounds, other factors such as
stem-cell-specific proteins appear to play roles in the entry
and retainment of CDy1, rendering it stem-cell selective. A
more detailed mechanism remains to be elucidated.
Among the few fluorescent dyes used for stem-cell
staining[9] is Aldefluor, which employs a fluorescent substrate
BODIPY-aminoacetaldehyde for aldehyde dehydrogenase
(ALDH)1A1.[10] It has been used to identify and isolate
certain types of stem cells including hematopoietic, neural
and mammary stem cells as well as cancer stem cells. Because
whether or not Aldefluor stains ESC has not been known, we
compared the cell selectivity of CDy1 with Aldefluor and
observed that Aldefluor stains neither mESC nor a human
ESC (hESC). Reciprocally, CDy1 stained both mESC and
hESC but not the human lung cancer cell line H522 which is
In summary, we have developed a novel bioimaging
probe, CDy1, for ESC and iPSC detection. The experimental
results presented herein strongly demonstrate that CDy1 can
be used for the identification and isolation of live ESC and
iPSC without the aid of a genetic reporter system at an earlier
stage of the reprogramming and during the ESC differ-
entiation. To our knowledge, no ESC- or iPSC-selective
fluorescent probe has been reported yet. This new probe,
CDy1, would be a useful tool for stem-cell research.
Received: April 26, 2010
Revised: July 30, 2010
Published online: September 2, 2010
Keywords: cell recognition · flurorescence · imaging agents ·
.
screening · stem cells
Angew. Chem. Int. Ed. 2010, 49, 7497 –7500
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7499

Communications
[5] J. S. Lee, Y. K. Kim, M. Vendrell, Y. T. Chang, Mol. Biosyst.
2009, 5, 411.
[6] Y. H. Ahn, J. S. Lee, Y. T. Chang, J. Am. Chem. Soc. 2007, 129,
4510.
[1] I. H. Park, N. Arora, H. Huo, N. Maherali, T. Ahfeldt, A.
Shimamura, M. W. Lensch, C. Cowan, K. Hochedlinger, G. Q.
Daley, Cell 2008, 134, 877.
[2] J. T. Dimos, K. T. Rodolfa, K. K. Niakan, L. M. Weisenthal, H.
Mitsumoto, W. Chung, G. F. Croft, G. Saphier, R. Leibel, R.
Goland, H. Wichterle, C. E. Henderson, K. Eggan, Science 2008,
321, 1218.
[3] N. Amariglio, A. Hirshberg, B. W. Scheithauer, Y. Cohen, R.
Loewenthal, L. Trakhtenbrot, N. Paz, M. Koren-Michowitz, D.
Waldman, L. Leider-Trejo, A. Toren, S. Constantini, G. Rechavi,
PLoS Med. 2009, 6, e29.
[7] Y. K. Kim, H. H. Ha, J. S. Lee, X. Bi, Y. H. Ahn, S. Hajar, J. J.
Lee, Y. T. Chang, J. Am. Chem. Soc. 2010, 132, 576.
[8] K. Takahashi, S. Yamanaka, Cell 2006, 126, 663.
[9] J. M. Zijlmans, J. W. Visser, K. Kleiverda, P. M. Kluin, R.
Willemze, W. E. Fibbe, Proc. Natl. Acad. Sci. USA 1995, 92, 8901.
[10] R. W. Storms, A. P. Trujillo, J. B. Springer, L. Shah, O. M. Colvin,
S. M. Ludeman, C. Smith, Proc. Natl. Acad. Sci. USA 1999, 96,
9118.
[4] E. J. Sutton, T. D. Henning, B. J. Pichler, C. Bremer, H. E.
Daldrup-Link, Eur. Radiol. 2008, 18, 2021.
[11] J. S. Moreb, J. R. Zucali, B. Ostmark, N. A. Benson, Cytometry
Part B 2007, 72, 281.
7500 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7497 –7500